"D“The people who are most needed in the contact center tomorrow are precisely the people who are least willing to have their toilet visit registered as an exception.” Merijn te Booij is surprised by the decisive influence of the work schedule in customer contact.

 

I grew up in the Netherlands. Grew up with Queen's Day on April 30th. Year after year. Orange everywhere. Flea markets on every street corner. An entire country collectively deciding that functioning normally is optional today. April 30th. Unwavering. Sacred. As self-evident as gravity.

Then, in 2013, we got a king.

King's Day. April 27th. The same national chaos, a different date. Simple in theory. Confusing in practice. I, after three decades of a deeply ingrained reflex, simply refused to cooperate. You can know that the date has changed. You can set three reminders in your calendar. And yet, every year on April 27th, you are a little surprised. King's Day is also a strange word. It is like being asked to forget how to ride a bicycle.

That is exactly what is happening to WFM professionals right now. Only they have been riding the same bike for thirty years.

For decades the beating heart of every contact center

The schedule has been the beating heart of the contact center for decades. Erlang C, interval scheduling, adherence rates, real-time deviation alerts; an entire field built around one obsession: knowing exactly where every employee needs to be at any given moment, and making sure they feel guilty if they fail to do so.

The tools have certainly evolved. Paper spreadsheets became cloud platforms. Dashboards became more aesthetically pleasing. AI was tacked onto them. Every major vendor, the well-known names with the gleaming booths at every contact center conference, has a roadmap full of promising innovations. Without exception, this is a shinier version of what they sold in 2015. The core idea—that the roster is the sacred centerpiece, to be optimized and maintained with the energy of a Special Enforcement Officer—has not shifted a millimeter. Sub-optimization remains sub-optimization. The model still collapses if one or two people suddenly do something different.

Thirty years of evolution. The same bike.

People are no longer the only shock absorber in customer contact.

When a peak occurs on Tuesday morning at 11:47, you no longer have to panic and pull employees from their lunch breaks to hope for the best. You have a whole portfolio of buffers. AI handles the routine volume. Callback requests redistribute the time pressure. Asynchronous channels delay handling. Gig employees chat. By the time a customer gets a human on the line, it is because the conversation really calls for it, not because the queue got too long.

If AI and 'asynchronous' processes handle the variation, accurate human interval scheduling becomes much less critical. You don't need seventeen employees at 10:45 and nineteen at 11:00. You need a capable pool of people available within a general time window. That is a fundamentally different, and much more manageable, planning issue. In short, the roster is not being outcompeted; it is being rendered irrelevant by the surrounding infrastructure.

The real price, therefore, is forecasting. Not the tactical four-week forecast that drives the following week's schedules, a month before it becomes actual. It becomes much more strategic. Demand intelligence that understands the real factors driving contact: product launches, billing cycles, service issues, changes in customer satisfaction, marketing campaigns, external events, errors, intent paths, data breaches. Yes, I am an Odigo customer.

Forecasting that asks not just “how many calls at 11 o’clock?” but “what do our customers need over the next ninety days, and is our human/AI capacity mix ready for that?”

Demand intelligence

True demand intelligence does not stop at volume and occupancy. It extends into areas the rest of the company has always cared about but was never invited to consider by WFM: revenue impact, customer sentiment, and service outcomes—expected problem areas. Why do contacts increase every third week of the month? Probably a billing issue. That is a product issue. Why does sentiment drop among customers who have contacted us more than twice? That is a churn signal and a reason for a conversation with Sales and Customer Success. Why do certain contact types reliably predict an upsell opportunity? That is commercial intelligence. It belongs in a commercial forecast.

When executed well, demand intelligence becomes a company-wide asset – a live signal regarding customer health, business performance, and operational risk that every department should be able to read. The contact center possesses an extraordinary amount of behavioral data. Currently, most of this is being used to justify the headcount for next month. That is a spectacular waste.

This is where the competitive advantage lies. Not in squeezing 98% adherence out of a team, something driven into silent frustration by micromanagement. But in understanding what your customers are telling you – before they tell your competitor.

The employees? They are done with the ankle bracelet model.

The WFM function of the future is not a scheduling department.

The WFM function of the future is not a scheduling department. It is a demand intelligence capability with a seat at the table – not only with operations, but also with finance, product, and commercial strategy. And significantly more interesting than staring at a red dot on an adherence dashboard.

And the employees? They are done with the ankle bracelet model.

The people most needed in the contact center tomorrow—capable of the complex, emotionally charged conversations that no AI can handle for the time being—are precisely the people least willing to register their bathroom break as an exception. They have choices. They make those choices. A 30 to 45 percent annual turnover is no mystery. It is a predictable reaction to a work environment designed around control rather than craftsmanship.

The alternative is a knowledge worker model. Not 'here is your shift,' but 'here is a time window in which your expertise is needed; indicate when it suits you.' The employee chooses their availability. The platform matches that availability to the forecast. AI and asynchronous processes fill the gaps. The roster does not disappear: it simply ceases to be the central hub around which everything revolves.

The roster won't disappear overnight. But its grip is loosening—torn loose by agentic AI, asynchronous flexibility, and a workforce that has collectively decided things can be done differently. The end of the roster is not the end of Workforce Management. It is, however, a redesign that has been delayed for too long. April 27th is actually fine. You just have to be willing to update the calendar.

(Ziptone/Merijn te Booij)

Featured, Human Resources, Opinions